US8358871B2 - Method and device for detecting and correcting skewed image data - Google Patents

Method and device for detecting and correcting skewed image data Download PDF

Info

Publication number
US8358871B2
US8358871B2 US12/473,554 US47355409A US8358871B2 US 8358871 B2 US8358871 B2 US 8358871B2 US 47355409 A US47355409 A US 47355409A US 8358871 B2 US8358871 B2 US 8358871B2
Authority
US
United States
Prior art keywords
image data
rotating
binary
detecting
skewed
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active, expires
Application number
US12/473,554
Other languages
English (en)
Other versions
US20100215285A1 (en
Inventor
Chien-Hui Tu
Cheng-Yueh Lo
De-Wei Huang
Yung-Hsi Wu
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Aver Information Inc
Original Assignee
Avermedia Information Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Avermedia Information Inc filed Critical Avermedia Information Inc
Assigned to AVERMEDIA INFORMATION, INC. reassignment AVERMEDIA INFORMATION, INC. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: WU, YUNG-HSI, HUANG, De-wei, Lo, Cheng-Yueh, Tu, Chien-Hui
Publication of US20100215285A1 publication Critical patent/US20100215285A1/en
Assigned to AVER INFORMATION INC. reassignment AVER INFORMATION INC. CHANGE OF NAME (SEE DOCUMENT FOR DETAILS). Assignors: AVERMEDIA INFORMATION, INC.
Application granted granted Critical
Publication of US8358871B2 publication Critical patent/US8358871B2/en
Active legal-status Critical Current
Adjusted expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/24—Aligning, centring, orientation detection or correction of the image
    • G06V10/242—Aligning, centring, orientation detection or correction of the image by image rotation, e.g. by 90 degrees
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/24—Aligning, centring, orientation detection or correction of the image
    • G06V10/243—Aligning, centring, orientation detection or correction of the image by compensating for image skew or non-uniform image deformations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G06T2207/10008—Still image; Photographic image from scanner, fax or copier
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20048—Transform domain processing
    • G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20068—Projection on vertical or horizontal image axis

Definitions

  • the present invention relates to a method and a device for detecting and correcting skewed image data, and more particularly to a method and a device for detecting and correcting skewed image data in an image processing apparatus.
  • the common image processing apparatuses include for example scanners, printers, copiers, facsimile machines, document cameras, and the like.
  • a skewed document detection and correction technology plays an important role in the field of document analysis systems.
  • the challenge of the skewed document detection and correction technology is to remove the non-text symbols (e.g. graphs) of the document.
  • the first approach usually needs a great quantity of memory capacity.
  • the first approach is only applicable when the text size and the image noise comply with a specified condition.
  • the text portion is transformed into multiple lines by computations, then the rotating angles of the text lines are calculated, and finally a skew angle of the document is estimated according to the rotating angles.
  • the second approach removes the non-text symbols (e.g. graphs) of the document while ignoring the document contents. According to four corners or boundaries realized from the color difference between the document and the background, the skew degree is obtained. If the color difference is not evident or the boundary is beyond the scanning range, the second approach is not applicable.
  • the prior art technologies need a great quantity of memory capacity to store the sorted and statistic data and thus are not suitably implemented by hardware components.
  • a skewed image data detecting and correcting method for use in an image processing apparatus.
  • the skewed image data detecting and correcting method includes the following steps. First of all, an image data is received and a binary digitizing operation is performed on the image data to obtain a binary image data, wherein each pixel of the binary image data is expressed by a single bit. Then, the binary image data is rotated by multiple different rotating angles, thereby obtaining multiple rotated binary image data. Then, pixel numbers of all horizontal rows of the rotated binary image data are totalized, thereby obtaining corresponding multiple horizontal pixel number distribution curves.
  • a high-pass filtering procedure is performed to filter off low-frequency noise contained in the horizontal pixel number distribution curves, thereby obtaining corresponding multiple high-frequency signal curves.
  • the square sums of respective high-frequency signal curves are calculated, thereby obtaining corresponding multiple index values.
  • a rotating correction operation is performed on the image data according to the rotating angle corresponding to a maximum of the index values, thereby obtaining a corrected image data.
  • the image data is obtained by scaling down an original image data.
  • the scaled-down image data is divided into multiple n ⁇ n cells, a brightness inversion operation is performed on all pixels of the n ⁇ n cells that have more than half of pixels are ranged from the medium level to the full black level, and the high-frequency noise at the borders of the n ⁇ n cells is deleted.
  • the borders of the n ⁇ n cells are reconstructed by interpolation or extrapolation with neighboring pixels.
  • the step of rotating the binary image data by multiple different rotating angles to obtain multiple rotated binary image data includes sub-steps of rotating the binary image data by a first rotating angle to obtain a first rotated binary image data, and rotating the binary image data by a second rotating angle to obtain a second rotated binary image data if the first rotating angle is within a searching angle range.
  • the step of calculating the square sums of respective high-frequency signal curves is implementing by the square sums of only the positive value portions of the high-frequency signal curves.
  • an inverse value of the rotating angle corresponding to the maximum of the index values denotes a skew angle of the image data.
  • a skewed image data detecting and correcting device of an image processing apparatus includes a skew angle detecting module and an image rotating correction module.
  • the skew angle detecting module implements the following procedures: receiving an image data and performing a binary digitizing operation on the image data to obtain a binary image data, rotating the binary image data by multiple different rotating angles to obtain multiple rotated binary image data, totalizing pixel numbers of all horizontal rows of the rotated binary image data to obtain corresponding multiple horizontal pixel number distribution curves, performing a high-pass filtering procedure to filter off low-frequency noise contained in the horizontal pixel number distribution curves to obtain corresponding multiple high-frequency signal curves, calculating the square sums of respective high-frequency signal curves to obtain corresponding multiple index values, and detecting a skew angle of the image data according to the rotating angle corresponding to a maximum of the index values.
  • the image rotating correction module is communicated with the skew angle detecting module for performing a rotating correction operation on the image data according to the skew angle, thereby obtaining a corrected image data.
  • the image data is obtained by scaling down an original image data.
  • the image rotating correction module performs the rotating correction operation on the original image data according to the skew angle, thereby obtaining a corrected original image data.
  • the skew angle detecting module includes a pre-processing module, a memory, a projection profile processor, a high-pass filter, and a statistics data collection and skew angel discriminator.
  • FIG. 1 is a schematic functional block diagram illustrating a skewed image data detecting and correcting device according to an embodiment of the present invention
  • FIG. 2 schematically illustrates a flowchart of a skewed image data detecting and correcting method according to the present invention
  • FIG. 3A schematically illustrates a horizontal pixel number distribution curve obtained by totalizing pixel numbers of all horizontal rows of the binary image data of a non-skewed pure-text image
  • FIG. 3B schematically illustrates a horizontal pixel number distribution curve obtained by totalizing pixel numbers of all horizontal rows of the binary image data of a slightly skewed pure-text image
  • FIGS. 4A and 4B schematically illustrate high-frequency signal curves obtained by filtering off low-frequency noise contained in the horizontal pixel number distribution curves
  • FIG. 5 schematically illustrates the efficacy of performing the brightness inversion operation on the image data
  • FIG. 6 schematically illustrates a process of deleting the high-frequency noise at the borders of the n ⁇ n cells and reconstructing the blank regions by interpolation or extrapolation with the neighboring pixels.
  • FIG. 1 is a schematic functional block diagram illustrating a skewed image data detecting and correcting device according to an embodiment of the present invention.
  • the skewed image data detecting and correcting device 2 includes a skew angle detecting module 21 and an image rotation correcting module 22 .
  • a skew angle of the original image data 1 is realized.
  • the information associated with the skew angle is then transmitted from the angle detecting module 21 to the image rotating correction module 22 .
  • the image rotating correction module 22 performs a rotating correction operation on the original image data 1 , thereby obtaining a corrected image data.
  • the corrected image data is transmitted to the back-end display unit 31 for display or the storage unit 32 for storage.
  • the skew angle detecting module 21 comprises a pre-processing module 210 , a memory 211 , a projection profile processor 212 , a high-pass filter 213 , and a statistics data collection and skew angel discriminator 214 .
  • FIG. 2 schematically illustrates a flowchart of a skewed image data detecting and correcting method according to the present invention.
  • the skewed image data detecting and correcting method will be illustrated in more details with reference to FIGS. 1 and 2 .
  • an original image data 1 is received by the pre-processing module 210 and the pre-processing module 210 performs a scaling-down operation on the original image data 1 (Step 41 ).
  • the original image data 1 is scaled down to have a 256 ⁇ 192 pixel resolution in order to save the memory usage.
  • the procedure of performing the scaling-down operation may be omitted. Without the scaling-down operation, the memory usage is relatively larger and the computing load is heavier.
  • the pre-processing operation is a binary digitizing operation for processing the scaled-down image data into a binary image data.
  • each pixel is expressed by a single bit.
  • the binary image data is then stored in the memory 211 (Step 42 ).
  • the binary image data stored in the memory 211 is read out by the projection profile processor 212 .
  • the pixel numbers (0 or 1) of all horizontal rows of the binary image data are respectively totalized by the projection profile processor 212 , thereby obtaining a horizontal pixel number distribution curve.
  • the low-frequency noise contained in the horizontal pixel number distribution curve is filtered off by the high-pass filter 213 , thereby obtaining a high-frequency signal curve (Step 43 ).
  • the square sum of the high-frequency signal curve is calculated by the statistics data collection and skew angel discriminator 214 , thereby obtaining an index value (Step 44 ).
  • the binary image data stored in the memory 211 is read out by the statistics data collection and skew angel discriminator 214 .
  • the binary image data is rotated by a specified rotating angle, thereby obtaining a rotated binary image data (Step 45 ). If the specified rotating angle falls into a searching angle range (Step 46 ), the Step 43 and the Step 44 are repeatedly done and thus another index value corresponding to the specified rotating angle is obtained. Until the specified rotating angle is beyond the searching angle range (Step 46 ), a maximum index value is selected from all of these index values.
  • the inverse value of the rotating angle corresponding to the maximum index value denotes the skew angle of the image data (Step 47 ).
  • the information associated with the skew angle is then transmitted from the angle detecting module 21 to the image rotating correction module 22 .
  • the image rotating correction module 22 performs a rotating correction operation on the original image data 1 , thereby obtaining a corrected image data.
  • the corrected image data is transmitted to the back-end display unit 31 for display or the storage unit 32 for storage.
  • the skewed image data detecting and correcting device 2 is further communicated with an application program interface 30 . Via the application program interface 30 , the specified rotating angle and the searching angle range could be preset in the statistics data collection and skew angel discriminator 214 .
  • the skewed image data detecting and correcting method of the present invention tries to emphasize the text characteristics, eliminate some noise, and reduce the interference from the non-text symbols.
  • a brightness inversion operation is performed before the step of processing the scaled-down image data into the binary image data (Step 42 ).
  • the gray levels of the pixels included in the scaled-down image data are ranged from full white (e.g. level 0) to full black (e.g. level 255).
  • the brightness inversion operation is carried out. For example, if at least 32 pixels in an 8 ⁇ 8 cell are ranged from the medium level (e.g. 128) to the full black level (e.g. 255), the brightness inversion operation is performed on all pixels in the 8 ⁇ 8 cell. Due to the brightness inversion operation, the erroneous discrimination is minimized or eliminated and the correction accuracy is enhanced.
  • FIG. 5 schematically illustrates the efficacy of performing the brightness inversion operation on the image data.
  • a white-text-on-black-background image 51 is directly subject to a binary digitizing operation
  • a binary image 52 is obtained.
  • the text symbols and the non-text symbols e.g. graphs
  • the 8 ⁇ 8 cell 53 is subject to a brightness inversion operation
  • a brightness-inversed image 54 is obtained.
  • the brightness-inversed image 54 is then subject to a binary digitizing operation, thereby obtaining another binary image 55 .
  • the text characteristics of the binary image 55 are emphasized and thus the possibility of causing erroneous discrimination is reduced.
  • n ⁇ n cell could be altered as required or according to the scaled-down image data.
  • the brightness inversion operation may produce undesirable high-frequency noise in the binary image because the brightness values of the neighboring cells are very distinct from each other.
  • saw-toothed lines 600 are possibly generated in the binary image 6 at the borders of the n ⁇ n cells. If the high-frequency noise at the borders is directly deleted, grid-like blank regions 601 are possibly generated in the binary image 6 . By interpolation or extrapolation with the neighboring pixels, proper pixel numbers of the blank regions 601 of an 8 ⁇ 8 cell 61 are reconstructed. As such, the high-frequency noise included in the reconstructed image 62 is minimized.
  • FIG. 3A schematically illustrates a horizontal pixel number distribution curve obtained by totalizing pixel numbers of all horizontal rows of the binary image data of a non-skewed pure-text image.
  • FIG. 3B schematically illustrates a horizontal pixel number distribution curve obtained by totalizing pixel numbers of all horizontal rows of the binary image data of a slightly skewed pure-text image.
  • the horizontal projection profile peak value (e.g. 60) of the non-skewed pure-text image is greater than the horizontal projection profile peak value (e.g. 40) of the slightly skewed pure-text image.
  • the horizontal projection profile peak value is effective to distinguish the non-skewed pure-text image from the slightly skewed pure-text image, there is still a drawback. That is, if the image contains both text symbols and non-text symbols, the efficacy of using the horizontal projection profile peak value to discriminate the skewed amount is largely reduced. The inventor also fount that the peak and trough of the horizontal projection profile are very distinguished.
  • the low-frequency noise contained in the horizontal pixel number distribution curve is filtered off by the high-pass filter 213 , thereby obtaining a high-frequency signal curve.
  • the horizontal projection profile of the skewed document or the horizontal projection profile of non-text symbols (e.g. graphs) of the document is declined.
  • An example of the high-pass filter 213 includes but is not limited to a finite impulse response (FIR) filter.
  • the high-frequency signal curve is transmitted to the statistics data collection and skew angel discriminator 214 .
  • the square sum of the high-frequency signal curve is calculated by the statistics data collection and skew angel discriminator 214 , thereby obtaining an index value.
  • the square sum of only the positive value portion of the high-frequency signal curve is calculated by the statistics data collection and skew angel discriminator 214 in order to increase the weight of the peaks.
  • the binary image data is rotated by multiple different rotating angle within the within a searching angle range.
  • a two-stage searching process is performed.
  • the binary image data is rotated by 5 degrees each time within a searching angle range between ⁇ 45 degrees and +45 degrees, thereby obtaining multiple rotated binary image data.
  • multiple index values corresponding to the rotating angles are obtained.
  • a maximum index value corresponding to a first-stage rotating angle is selected from all of these index values.
  • the binary image data is rotated by 1 degree each time within another searching angle range between ⁇ 5 degrees and +5 degrees.
  • multiple index values corresponding to the rotating angles are obtained.
  • a maximum index value corresponding to a second-stage rotating angle is selected from all of these index values.
  • the skew angle of the skewed document could be precisely detected.
  • the information associated with the skew angle is then transmitted from the angle detecting module 21 to the image rotating correction module 22 .
  • the image rotating correction module 22 performs a rotating correction operation on the original image data 1 , thereby obtaining a corrected image data.
  • the corrected image data is transmitted to the back-end display unit 31 for display or the storage unit 32 for storage.
  • the present invention is effective for detecting and correcting skewed image data in an image processing apparatus.
  • the image processing apparatus includes for example a scanner, a printer, a copier, a facsimile machine, a document camera, and the like.
  • the skewed image data detecting and correcting method of the present invention can detect and correct the skewed image by using reduced hardware resource without the need of removing the non-text symbols.
  • the device and method of the present invention could be implemented by hardware components, the memory capacity is saved to store the sorted and statistic data.
  • the skewed image data detecting and correcting method could be applied to any photoelectric system having the image processing apparatus.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Processing (AREA)
  • Editing Of Facsimile Originals (AREA)
US12/473,554 2009-02-20 2009-05-28 Method and device for detecting and correcting skewed image data Active 2031-09-23 US8358871B2 (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
TW098105465 2009-02-20
TW98105465A 2009-02-20
TW098105465A TWI425444B (zh) 2009-02-20 2009-02-20 影像資料歪斜偵測校正方法與裝置

Publications (2)

Publication Number Publication Date
US20100215285A1 US20100215285A1 (en) 2010-08-26
US8358871B2 true US8358871B2 (en) 2013-01-22

Family

ID=42109990

Family Applications (1)

Application Number Title Priority Date Filing Date
US12/473,554 Active 2031-09-23 US8358871B2 (en) 2009-02-20 2009-05-28 Method and device for detecting and correcting skewed image data

Country Status (4)

Country Link
US (1) US8358871B2 (de)
EP (1) EP2221767B1 (de)
JP (1) JP4824101B2 (de)
TW (1) TWI425444B (de)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140044353A1 (en) * 2010-03-22 2014-02-13 Sharp Laboratories Of America, Inc. Methods and Systems for Estimation of Document Skew in an Image
US9288362B2 (en) * 2014-02-03 2016-03-15 King Fahd University Of Petroleum And Minerals Technique for skew detection of printed arabic documents
US9621761B1 (en) 2015-10-08 2017-04-11 International Business Machines Corporation Automatic correction of skewing of digital images
US10582085B2 (en) * 2016-07-12 2020-03-03 Zhuhai Seine Technology Co., Ltd. Method and device for image processing, and apparatus for image formation

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20110038223A (ko) * 2009-10-08 2011-04-14 삼성전자주식회사 스캐너 및 화상형성장치와, 스캐너의 화상 보정 방법
TWI543110B (zh) 2015-04-27 2016-07-21 虹光精密工業股份有限公司 產生影像之方法及影像處理設備
CN105095896B (zh) * 2015-07-29 2019-01-08 江苏邦融微电子有限公司 一种基于查找表的图像畸变校正方法
CN105426887B (zh) * 2015-10-30 2019-04-12 北京奇艺世纪科技有限公司 一种文本图像校正的方法及装置
JP6999318B2 (ja) * 2017-07-24 2022-01-18 ラピスセミコンダクタ株式会社 撮像装置及び水平方向検出方法
CN111950554A (zh) * 2020-08-17 2020-11-17 深圳市丰巢网络技术有限公司 一种身份证识别方法、装置、设备及存储介质
CN113420762B (zh) * 2021-07-01 2024-02-06 京东科技控股股份有限公司 一种图像处理方法、系统、电子设备及存储介质
TWI810813B (zh) * 2022-02-11 2023-08-01 圓展科技股份有限公司 影像校正系統及方法

Citations (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5521985A (en) * 1992-08-13 1996-05-28 International Business Machines Corporation Apparatus for recognizing machine generated or handprinted text
JPH11331558A (ja) 1998-05-21 1999-11-30 Dainippon Screen Mfg Co Ltd ページ位置補正方法、装置および記録媒体
US6064778A (en) 1993-10-25 2000-05-16 Primax Electronics Ltd. Method and apparatus for near real-time document skew compensation
US6430320B1 (en) 1998-04-09 2002-08-06 Hewlett-Packard Company Image processing system with automatic image cropping and skew correction
US20020131642A1 (en) * 2001-02-02 2002-09-19 Lee Shih-Jong J. Robust method for automatic reading of skewed, rotated or partially obscured characters
US20030128895A1 (en) * 2002-01-07 2003-07-10 Xerox Corporation Parallel non-iterative method of determining and correcting image skew
JP2004128643A (ja) 2002-09-30 2004-04-22 Matsushita Electric Ind Co Ltd 画像の傾き補正方法
US6771842B1 (en) 1998-05-28 2004-08-03 Fujitsu Limited Document image skew detection method
US20050281483A1 (en) * 2004-06-08 2005-12-22 Benq Corporation Tilted angle detection for document image deskew
US7027666B2 (en) 2002-10-01 2006-04-11 Eastman Kodak Company Method for determining skew angle and location of a document in an over-scanned image
US7110568B2 (en) 2000-06-27 2006-09-19 Solystic Segmentation of a postal object digital image by Hough transform
US7133573B2 (en) 2002-11-26 2006-11-07 Eastman Kodak Company Document skew accommodation
US7200285B2 (en) 2004-08-21 2007-04-03 Xerox Corporation Detecting skew angle in a scanned image
US7277600B2 (en) 2002-08-27 2007-10-02 Oce Print Logic Technologies S.A. Skew detection
US7336813B2 (en) 2004-04-26 2008-02-26 International Business Machines Corporation System and method of determining image skew using connected components
US20090316987A1 (en) * 2008-06-23 2009-12-24 Bing Ouyang System and Method for Determining High Frequency Content in an Analog Image Source
US20100238356A1 (en) * 2009-03-18 2010-09-23 Victor Company Of Japan, Ltd. Video signal processing method and apparatus

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5557784A (en) * 1995-03-30 1996-09-17 International Business Machines Corporation Power on timer for a personal computer system
US7213140B2 (en) * 2003-10-30 2007-05-01 Micro-Star Int'l Co., Ltd. Method for self-starting a computer
JP4556813B2 (ja) * 2005-09-08 2010-10-06 カシオ計算機株式会社 画像処理装置、及びプログラム
US7752382B2 (en) * 2005-09-09 2010-07-06 Sandisk Il Ltd Flash memory storage system and method
TW200727126A (en) * 2006-01-12 2007-07-16 Mitac Int Corp Power management method

Patent Citations (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5521985A (en) * 1992-08-13 1996-05-28 International Business Machines Corporation Apparatus for recognizing machine generated or handprinted text
US6064778A (en) 1993-10-25 2000-05-16 Primax Electronics Ltd. Method and apparatus for near real-time document skew compensation
US6430320B1 (en) 1998-04-09 2002-08-06 Hewlett-Packard Company Image processing system with automatic image cropping and skew correction
JPH11331558A (ja) 1998-05-21 1999-11-30 Dainippon Screen Mfg Co Ltd ページ位置補正方法、装置および記録媒体
US6771842B1 (en) 1998-05-28 2004-08-03 Fujitsu Limited Document image skew detection method
US7110568B2 (en) 2000-06-27 2006-09-19 Solystic Segmentation of a postal object digital image by Hough transform
US20020131642A1 (en) * 2001-02-02 2002-09-19 Lee Shih-Jong J. Robust method for automatic reading of skewed, rotated or partially obscured characters
US6856697B2 (en) * 2001-02-02 2005-02-15 Shih-Jong J. Lee Robust method for automatic reading of skewed, rotated or partially obscured characters
US6985640B2 (en) 2002-01-07 2006-01-10 Xerox Corporation Parallel non-iterative method of determining and correcting image skew
US20030128895A1 (en) * 2002-01-07 2003-07-10 Xerox Corporation Parallel non-iterative method of determining and correcting image skew
US7277600B2 (en) 2002-08-27 2007-10-02 Oce Print Logic Technologies S.A. Skew detection
JP2004128643A (ja) 2002-09-30 2004-04-22 Matsushita Electric Ind Co Ltd 画像の傾き補正方法
US7027666B2 (en) 2002-10-01 2006-04-11 Eastman Kodak Company Method for determining skew angle and location of a document in an over-scanned image
US7133573B2 (en) 2002-11-26 2006-11-07 Eastman Kodak Company Document skew accommodation
US7336813B2 (en) 2004-04-26 2008-02-26 International Business Machines Corporation System and method of determining image skew using connected components
US20050281483A1 (en) * 2004-06-08 2005-12-22 Benq Corporation Tilted angle detection for document image deskew
US7200285B2 (en) 2004-08-21 2007-04-03 Xerox Corporation Detecting skew angle in a scanned image
US20090316987A1 (en) * 2008-06-23 2009-12-24 Bing Ouyang System and Method for Determining High Frequency Content in an Analog Image Source
US20100238356A1 (en) * 2009-03-18 2010-09-23 Victor Company Of Japan, Ltd. Video signal processing method and apparatus

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
Fahn et al. "A featureless image registration method for deskewed document images" Journal of Technology, vol. 21, No. 4, pp. 405-414, 2006.

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140044353A1 (en) * 2010-03-22 2014-02-13 Sharp Laboratories Of America, Inc. Methods and Systems for Estimation of Document Skew in an Image
US9098771B2 (en) * 2010-03-22 2015-08-04 Sharp Laboratories Of America, Inc. Methods and systems for estimation of document skew in an image
US9288362B2 (en) * 2014-02-03 2016-03-15 King Fahd University Of Petroleum And Minerals Technique for skew detection of printed arabic documents
US9621761B1 (en) 2015-10-08 2017-04-11 International Business Machines Corporation Automatic correction of skewing of digital images
US10176395B2 (en) 2015-10-08 2019-01-08 International Business Machines Corporation Automatic correction of skewing of digital images
US10582085B2 (en) * 2016-07-12 2020-03-03 Zhuhai Seine Technology Co., Ltd. Method and device for image processing, and apparatus for image formation

Also Published As

Publication number Publication date
TWI425444B (zh) 2014-02-01
JP4824101B2 (ja) 2011-11-30
EP2221767A1 (de) 2010-08-25
TW201032179A (en) 2010-09-01
US20100215285A1 (en) 2010-08-26
JP2010200288A (ja) 2010-09-09
EP2221767B1 (de) 2013-03-27

Similar Documents

Publication Publication Date Title
US8358871B2 (en) Method and device for detecting and correcting skewed image data
US8995012B2 (en) System for mobile image capture and processing of financial documents
US9509884B2 (en) Skew detection
KR100945809B1 (ko) 이미지의 이진화
US8582862B2 (en) Mobile image quality assurance in mobile document image processing applications
US7660483B2 (en) Method and apparatus for removing noise from a digital image
US20120274991A1 (en) System and method for document orientation detection
WO2018010547A1 (zh) 一种图像处理方法、装置及图像形成设备
JP6021665B2 (ja) 画像処理装置、画像処理方法及びコンピュータプログラム
US20110135362A1 (en) Image forming apparatus and method for de-skewing thereof
US8306335B2 (en) Method of analyzing digital document images
KR101011908B1 (ko) 디지털 이미지의 노이즈 감소 방법 및 이를 이용한 이미지 처리 장치
JP2003016440A5 (de)
US7545535B2 (en) Robust automatic page size detection algorithm for scan application
JP3989341B2 (ja) 画像処理装置
US9858512B2 (en) Image processing apparatus with an improved detection of ruled lines
JP5480102B2 (ja) 画像処理装置、画像形成装置、および画像処理方法
JP2009284190A (ja) 画像読み取り装置および画像読み取りシステム
JP5482945B1 (ja) 画像処理装置及びプログラム
JP2947126B2 (ja) 原稿イメージ取り込み装置
JP6810892B2 (ja) 画像処理装置
Hu et al. Determining Optimal Resolution for Scanned Document Raster Content
JP3962611B2 (ja) 画像処理装置およびその方法
JP3436003B2 (ja) 画像処理装置
JPH0832802A (ja) 画像処理装置

Legal Events

Date Code Title Description
AS Assignment

Owner name: AVERMEDIA INFORMATION, INC., TAIWAN

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:TU, CHIEN-HUI;LO, CHENG-YUEH;HUANG, DE-WEI;AND OTHERS;SIGNING DATES FROM 20090507 TO 20090508;REEL/FRAME:022745/0992

AS Assignment

Owner name: AVER INFORMATION INC., TAIWAN

Free format text: CHANGE OF NAME;ASSIGNOR:AVERMEDIA INFORMATION, INC.;REEL/FRAME:027439/0087

Effective date: 20080110

STCF Information on status: patent grant

Free format text: PATENTED CASE

FPAY Fee payment

Year of fee payment: 4

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 8TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1552); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

Year of fee payment: 8